Defacing Evaluation Pipeline

---

Setup

With Pip virtualenv

• Step 0: Set up virtual environment

  > pip3 install virtualenv # or sudo apt install python3-virtualenv
  > virtualenv .venv
  > source .venv/bin/activate
  

• Install cmake

  > sudo apt update
  > sudo apt install cmake
  

• (If necessary) Install tkinter via system packages

  > sudo apt-get install python3-tk
  

• Install Python dependencies

  > pip3 install -r requirements.txt
  

  [!IMPORTANT]

• Configure XNAT access (only needed for data import) Create a .env file in the src/ folder with the following variables: > XNAT_ADDRESS > XNAT_ALIAS > XNAT_SECRET > XNAT_PROJECT

With Miniconda

Install Miniconda

Linux - see Quickinstallation from Miniconda

  > mkdir -p ~/miniconda3
  > wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -O ~/miniconda3/miniconda.sh
  > bash ~/miniconda3/miniconda.sh -b -u -p ~/miniconda3
  > rm ~/miniconda3/miniconda.sh
  > source ~/miniconda3/bin/activate
  > conda init --all

Create and activate environment

  > conda create --name defacing
  > conda activate defacing
  > conda install python=3.12

Install requirements

  > cd defacing_evaluation/
  > pip install -r requirements.txt
  > pip install --upgrade pyvista

Data import

Run the following commands from the repository folder.

Generation overview CSV

• First use:

    > python3 src/main.py step1_overview
  

• Continued use with existing (unfinished) csv file:

  [!TIP] If you run the code again on the same day, you can also omit the filename of the csv.

    > python3 src/main.py step1_overview --csv_overviewfile path_to_unfinished_csv_overview.csv
  

• Use with your own dicom tag csv file:

  First use:

      > python3 src/main.py step1_overview --csv_dicomtags individual_dicom_taglist.csv 
  

  Continued use with unfinished csv file:

      > python3 src/main.py step1_overview --csv_dicomtags individual_dicom_taglist.csv --csv_overviewfile path_to_unfinished_csv_overview.csv
  

Caution: The CSV uses the pipe (|) separator

Import data

> python3 src/main.py step1_import --overview_csv specify_overview_file_path.csv

Defacing

> python3 src/main.py step2

Evaluation

Collect features

> python3 src/main.py step3_1

Manual groundtruth labelling

> python3 src/main.py step3_2

Machine Learning Modelling

• Create final dataset from evaluation and feature dataframes

  > python3 src/main.py step4_final_dataset
  

• Inter-Rater Reliability

  > python3 src/main.py step4_interrater
  

• MCCV nested CV

  > python3 src/main.py step4_nested_cv
  

Visuals

Visual scripts are not integrated in main.py. Please run them individually as needed.

Troubleshooting

• EOF Errors: If you encounter an EOFError during data loading, the file may be corrupted. The safest approach is to delete the file and reimport it from the source (e.g., XNAT).
• Dependencies: Some dependencies (e.g. tkinter) may require system-level installation or upgrades.